Performance regression?

[mcfarljm]

I'm seeing a performance regression in steepest descent in Ocean 4.1 vs Ocean 3.5. With Ocean 3.5, the sample code runs in about 2.4s, whereas with Ocean 4.1 it runs in about 3.9s. Is this expected? From what I can tell, they're both using the same version of greedy, so it seems surprising.

from time import perf_counter

import dimod
import greedy

bqm = dimod.generators.anti_crossing_clique(150)
sampler = greedy.SteepestDescentSolver()

def solve(bqm, sampler):
    results = sampler.sample(bqm, num_reads=50000)
    return results.first

tic = perf_counter()
first = solve(bqm, sampler)
toc = perf_counter()

print(toc-tic)

(Use of anti_crossing_clique as the test BQM is not significant, I just wanted something deterministic. I also saw similar differences with randomly generated BQM's.)

[randomir]

I've hunt this down to slow energy calculation on the new default BQM (in dimod 0.10/Ocean 4).

See dwavesystems/dimod#1025.

The irony is the energies are calculated during initial states generation (in dimod.Initialized.parse_initial_states, i.e. SampleSet.from_samples_bqm), for randomly generated states, only to be discarded by the sampler :-/

[arcondello]

In the long run we'll want to make having an energy data vector in sample sets optional. But for right now this is one of those prices of abstraction.

[randomir]

Absolutely. We've being toying with this idea for a while now, but with CQM out, we might want to prioritize it.

[mcfarljm]

Appears that this has been resolved in Ocean 5.0.